CN108664472B - Natural language processing method, device and equipment - Google Patents
Natural language processing method, device and equipment Download PDFInfo
- Publication number
- CN108664472B CN108664472B CN201810434229.4A CN201810434229A CN108664472B CN 108664472 B CN108664472 B CN 108664472B CN 201810434229 A CN201810434229 A CN 201810434229A CN 108664472 B CN108664472 B CN 108664472B
- Authority
- CN
- China
- Prior art keywords
- information
- user
- intention
- corpus
- natural language
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a natural language processing method, a natural language processing device and natural language processing equipment, wherein the method comprises the following steps: acquiring information currently input by a user, and detecting whether the information has a missing intention object and/or an intention purpose; if the information is detected to have a missing intention object and/or intention purpose, calling an exclusive corpus corresponding to the user to acquire historical corpus information of the user; performing natural language analysis on historical corpus information of a user, and extracting filling information matched with missing intention objects and/or intention purposes; and combining the filling information and the currently input information into complete target information, and performing corresponding processing according to an intention object and an intention purpose in the target information. Therefore, the integration of the speech data with uncertain expression and thinking is realized, the response efficiency of man-machine interaction can be improved, and the user experience of the man-machine interaction is improved.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a natural language processing method, apparatus, and device.
Background
Today, information is rapidly expanding, and natural language understanding technology has become the focus of people's daily life and science and technology industry attention, and is an important index for measuring machine intelligence. With the development of machine learning techniques and the popularization of deep learning techniques in recent years, natural language techniques have also begun to be applied to a plurality of fields (e.g., question answering, interactive robots, etc.), and the comprehension ability of machines to natural language has been improved, thereby lowering the threshold of man-machine communication.
In the related art, in the process of human-computer interaction, in the prior art, when information input by a user is ambiguous, a single question method which may be consulted by the user is mainly enumerated, and is slightly converted, such as conversion of sentence patterns, modifiers and the like, once the intention body and intention of the user in dialog information input by the user are ambiguous, the corresponding processing cannot be performed according to the sentence input by the user, at this time, the user can input the dialog information again, and the human-computer interaction experience of the user is not ideal.
Disclosure of Invention
The embodiment of the invention provides a natural language processing method, a natural language processing device and natural language processing equipment, and aims to solve the technical problems that in the prior art, when an intention subject and intention in dialog information currently input by a user are ambiguous in a human-computer interaction process, corresponding processing cannot be performed according to statements input by the user, the user is required to input the statements with definite meanings again, and the human-computer interaction experience of the user is not ideal.
The embodiment of the invention provides a natural language processing method, which comprises the following steps: acquiring information currently input by a user, and detecting whether the information has a missing intention object and/or an intention purpose; if the information is detected to have a missing intention object and/or intention purpose, calling an exclusive corpus corresponding to the user to acquire historical corpus information of the user; performing natural language analysis on the historical corpus information of the user, and extracting filling information matched with the missing intention object and/or intention purpose; and combining the filling information and the currently input information into complete target information, and performing corresponding processing according to an intention object and an intention purpose in the target information.
Another embodiment of the present invention provides a natural language processing apparatus, including: the detection module is used for acquiring the information currently input by a user and detecting whether the information has a missing intention object and/or an intention purpose; the acquisition module is used for calling an exclusive corpus corresponding to the user to acquire historical corpus information of the user if the information is detected to have a missing intention object and/or intention purpose; the extraction module is used for carrying out natural language analysis on the historical corpus information of the user and extracting filling information matched with the missing intention object and/or intention purpose; and the processing module is used for combining the filling information and the currently input information into complete target information and carrying out corresponding processing according to an intention object and an intention purpose in the target information.
Another embodiment of the present invention provides a terminal device, including: a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor implements the natural language processing method according to the embodiment of the first aspect of the present invention when executing the computer program.
A further embodiment of the present invention provides a storage medium, configured to store an application program, where the application program is configured to execute the natural language processing method according to the first aspect of the present invention.
The technical scheme provided by the embodiment of the invention has the following beneficial effects:
the missing intention object and/or the intention purpose target information is generated by complementing the information of the missing intention object and/or the intention purpose through the historical corpus information in the exclusive corpus corresponding to the user, and corresponding processing is carried out according to the intention object and the intention purpose in the target information, so that the integration of speech data with uncertain expression and intention is realized based on the exclusive corpus of the user, the response efficiency of man-machine interaction can be improved, and the user experience degree of man-machine interaction is improved.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which,
FIG. 1 is a flow diagram of a natural language processing method according to one embodiment of the invention;
FIG. 2 is a flow diagram of a natural language processing method according to another embodiment of the invention;
FIG. 3a is a schematic diagram of a human-machine interface including information input by a user;
FIG. 3b is a schematic diagram of a human-machine interface including a reorganization problem;
FIG. 3c is a schematic diagram of a human-machine interface including fill response information;
FIG. 3d is a schematic diagram of a human-machine interface including results of responses;
FIG. 4a is a schematic diagram of a human-computer interaction interface corresponding to a voice input function of a smart phone;
FIG. 4b is a schematic diagram of a human-computer interaction interface including text information corresponding to voice information;
FIG. 4c is a schematic diagram of a human-machine interface including answer information for machine responses;
FIG. 5 is a schematic structural diagram of a natural language processing apparatus according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a natural language processing apparatus according to another embodiment of the present invention;
FIG. 7 is an interaction flow diagram of a natural language processing method according to one embodiment of the invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
A natural language processing method, apparatus, and device according to an embodiment of the present invention are described below with reference to the accompanying drawings.
In the prior art, in the process of human-computer interaction, when the intention body and the intention in the dialog information currently input by the user are ambiguous, the corresponding processing cannot be performed according to the sentence input by the user, and the user needs to input the sentence with definite meaning again, so that the human-computer interaction experience of the user is not ideal.
In order to solve the above problem, in the natural language processing method according to an embodiment of the present invention, when an intention object and/or an intention purpose existing in language information currently input by a user is detected, historical corpus information of the user is obtained through an exclusive corpus corresponding to the user, filling information matching the missing intention object and/or the intention purpose is determined according to the historical corpus information, the filling information and the currently input information are combined into complete target information, and corresponding processing is performed according to the intention object and the intention purpose in the target information, so that completion of the missing intention object and/or the intention purpose information is achieved through the historical corpus information in the exclusive corpus corresponding to the user, target information including the intention object and the intention purpose is generated, and corresponding processing is performed according to the intention object and the intention purpose in the target information, therefore, the integration of the speech data with uncertain expression and thinking is realized, the response efficiency of the human-computer interaction can be improved, and the user experience of the human-computer interaction is improved.
The natural language processing method according to the embodiment of the present invention will be described in detail with reference to the drawings and specific embodiments.
Fig. 1 is a flowchart of a natural language processing method according to an embodiment of the present invention, as shown in fig. 1, the natural language processing method including:
The intention object is a body to which the user inputs information to express the intention.
Wherein the intended purpose is the purpose that the user wants to achieve to input information.
It should be noted that the user may input information in various ways, for example, the user may input information in a text or voice manner.
It should be understood that, when the user inputs information in a text manner, the obtained information currently input by the user is text information, and when the user inputs information in a voice manner, the obtained information currently input by the user is voice information.
It should be noted that, in practical application, a user may select to input information in a text or voice manner according to a requirement, which is not limited in this embodiment.
It should be understood that, in different human-computer interaction scenarios, the intention object and the intention purpose expressed by the user through the input information are different, and the following examples are illustrated:
for example, when the information currently input by the user is "please call XXX", the intention object is "XXX" and the intention is "call XXX".
For another example, when the information currently input by the user is "please give an alarm clock with 9 points set", the intention object is "alarm clock" and the intention object is "set 9 points".
For example, when the information currently input by the user is "how to do a social security card", the intention object is "the social security card" and the intention purpose is "how to do".
For another example, when the information currently input by the user is "what i like", in which the intention object is "i", the intention purpose is "what i like".
The terminal device refers to a device having an artificial intelligence dialog system, for example, the terminal device may be a smart phone, a portable device, a robot, or the like having an artificial intelligence dialog system, and the implementation is not limited to the terminal device.
In particular, in practical applications, a user needs to perform human-computer interaction in many scenarios, for example, in a chat session, a question session, or a task session, information currently input by the user in a text or voice manner may be acquired.
It is understood that, in different application scenarios, whether information has a missing intention object and/or an intention purpose may be detected in different ways, and examples are as follows:
in a first mode
Whether the information has missing intention objects and/or intention purposes is detected through a preset intention analysis model.
Specifically, after the information currently input by the user is obtained, the information currently input by the user may be input into the preset intention analysis model, and whether the information has a missing intention object and/or an intention purpose is determined according to an output result of the preset intention analysis model.
Mode two
As an exemplary embodiment, when it is determined that the information currently input by the user is voice information, the voice information currently input by the user is converted into text information, semantic analysis is performed on the text information corresponding to the voice information, and it is determined whether there is a missing intention object and/or intention purpose in the voice information according to a result of the semantic analysis.
As a possible implementation manner, after obtaining the text information corresponding to the information, it may be determined whether the text information includes a preset intention type semantic, and if it is determined that the text information includes the preset intention type semantic, an object corresponding to the preset intention type semantic is obtained.
The object corresponding to the preset intention type semantic meaning is an intention object.
The intention type semantics refers to some sentences used for representing the intention of the user, for example, the intention type semantics can be "go xxx", "see xxx", "eat xxx", "like what", "how to do", "at that do", "how to say", and "call xxx", etc.
For example, when the information currently input by the user is converted into the text information of "how to do the social security card", the text information is subjected to semantic analysis, and the type semantic meaning "how to do" of the intent in the text information can be determined through the analysis, and the object corresponding to the "intent type semantic" is the "social security card", so that the intent object in the information can be determined to be the "social security card" and the corresponding intent object can be determined to be the "how to do".
As another example, when it is determined that the information currently input by the user is text information, semantic analysis may be directly performed on the information, and it may be determined whether there is a missing intention object and/or an intention purpose in the information according to a result of the semantic analysis.
Wherein, the exclusive language database is used for storing the historical corpus information of the user.
The historical corpus information comprises historical dialogue information of the current man-machine dialogue input by the user before current information is input and historical dialogue information of other man-machine dialogues of the user.
In different application scenarios, it is detected that there is an intention object with missing information, and/or the situation of the intention purpose is divided into a plurality of cases, for example, as follows:
first example
When it is detected that there is an intended object of information and there is an intended purpose of missing, there are two cases, one is that there is no intended purpose of information, and the other is that there is only an intended purpose of information ambiguity.
In the present example, for example, assuming that the information currently input by the user is "social security card", it is possible to detect that the intention object "social security card" is present in the information without an intention purpose.
For example, if the information currently input by the user is "the amount of the XXX bank card limit", it is possible to detect that the intention object "XXX bank" exists in the information, and the "limit" is classified into the amount of the deposit limit, the amount of the local transfer limit, the amount of the remote transfer limit, the amount of the daily transfer limit, and the like.
Second example
When it is detected that information has an intended purpose and a missing intended object exists, there are two cases, one is that the information does not have the intended object, and the other is that the information has only an intended object whose information is unclear. In the present example, for example, assuming that the information currently input by the user is "do where", it is possible to detect that there is an intention purpose "do where" in the information, and there is no intention object.
For another example, it is assumed that the information currently input by the user is "how long it takes to drive a vehicle from a unit to shenzhen university", it is detectable that there is an intention purpose "how long it takes to drive a vehicle from a unit to shenzhen university" in the information, and since it is not described what specific unit is in the information of the user, it is detectable that the "unit" in the information is an intention object whose information is unclear.
Third example
When it is detected that there are missing intended objects and intended purposes, four cases are distinguished:
first case
It is detected that the information is not present with the intended object and that the intended purpose is not present.
For example, assuming that the information currently input by the user is "good and troublesome", it may be detected that there is no intention object and no intention destination in the information currently input by the user.
Second case
It is detected that there is no intention object in the information, and only an intention object whose information is unclear exists.
For example, if the information currently input by the user is "amount limit", at this time, because what amount limit is not determined in the information, and the amount limit is divided into the amount limit for deposit, the amount limit for local transfer, the amount limit for remote transfer, the amount limit for daily transfer, and the like, it is possible to detect that there is no intention object in the information currently input by the user, and only there is "amount limit" for the intention purpose for which the information is not determined.
Third case
It is detected that the information is only an intended object whose information is ambiguous and that there is no intended purpose.
For another example, assuming that the information currently input by the user is "my bank card", the user only says my bank card, does not say what the purpose of the bank card is, and does not say which bank card the user is, and therefore, at this time, it can be detected that the information currently input by the user has no intention, and the "my bank card" is an intention object whose information is unclear.
Fourth case
It is detected that the information is only an intended object whose information is ambiguous and only an intended purpose whose information is ambiguous.
For another example, assuming that the information currently input by the user is "how much my bank card is limited", since the user only says my bank card and does not say which bank card the user is, the "limit" is divided into the cases of how much deposit limit, how much local transfer limit, how much remote transfer limit, and how much daily transfer limit, etc., it is possible to detect that there is an intention object "my bank card" whose information is unclear and "how much limit" whose information is unclear in the information currently input by the user.
When it is detected that there is a missing intent object and/or an intent destination in the information, in order to avoid making a request for information from the user, that is, inquiring the user again about the missing intent object and/or what the intent destination is to obtain the missing intent object and/or the intent destination, the embodiment may call a dedicated corpus corresponding to the user to obtain the user's history corpus information and fill the missing intent object and/or the intent destination according to the user's history corpus information.
It should be noted that, in different application scenarios, the dedicated corpus corresponding to the user may be called in multiple ways to obtain the historical corpus information of the user, and this embodiment provides several ways of calling the dedicated corpus corresponding to the user to obtain the historical corpus information of the user, which are illustrated as follows:
as an example, a login account of a user is obtained, and a dedicated corpus corresponding to the login account is called to obtain historical corpus information of the user.
In the example, the account login account is used for calling the exclusive corpus corresponding to the user, so that when different terminal devices can be adopted for man-machine conversation at any time and any place, the information which is input by the user and lacks the intention purpose and/or the intention object can be supplemented based on the exclusive corpus of the user.
As another example, voiceprint processing is performed on currently input information to extract user voiceprint features of a user, and a dedicated corpus corresponding to the user voiceprint features is called to obtain historical corpus information of the user.
And 103, performing natural language analysis on the historical corpus information of the user, and extracting filling information matched with the missing intention object and/or intention purpose.
It should be noted that, in a non-application scenario, the extraction of the intent object and/or the filling information matching the intent object are different, which is described below with reference to a specific embodiment.
For example, if the information input by the user in the current round of conversation is "which is a nearest seafloor scoop from home", it can be detected that the information has an intention purpose, which is to obtain the nearest seafloor scoop from home, "home" is an intention object with unclear information, and in order to provide the nearest seafloor scoop from home to the user, an exclusive corpus of the user can be obtained.
For another example, if the user inputs information "where" in the current round of dialog during the current man-machine dialog, and the user can detect that the information has the intention purpose "where" and lacks the intention object, the history corpus information obtained from the user's exclusive corpus is: how the social security card is handled can be determined by analyzing historical corpus information, and the missing intention object is the social security card. It should be noted that the historical corpus information "how do the social security card" is the historical information input by the user in the previous round of the current conversation.
And 104, combining the filling information and the currently input information into complete target information, and performing corresponding processing according to an intention object and an intention purpose in the target information.
In practical application, there are various scenes related to human-computer interaction, and the purpose of human-computer interaction performed by a user in different scenes is different, so that corresponding processing performed according to an intention object and an intention purpose in target information is different in different human-computer interaction scenes, and the following examples are illustrated:
in a task human-computer interaction scene, analyzing target information to obtain a processing instruction of a user, calling a related application program according to an intention object and an intention purpose in the target information to perform information processing, and obtaining a response result corresponding to the processing instruction to feed back to the user.
In this example, for example, the information currently input by the user is "help me to taxi to the nearest social security bureau from home", and it is assumed that the filling information corresponding to the intent object "home" with ambiguous information is the science and academic park, the target information obtained according to the filling information and the currently input information is "help me to taxi to the nearest social security bureau from the science and academic park", at this time, the address information of the social security bureau closest to the science and academic park can be obtained from the map application program, and the taxi-taxi software can taxi to the user according to the science and academic garden and the obtained address information of the social security bureau.
In a second example, in a question-answer interaction scenario, corresponding answer information is obtained according to an intention object and an intention purpose in target information, and the answer information is fed back to a user.
In this example, for example, the information currently input by the user is "how to do the social security card", it is detected that the intention object in the information is "the social security card", and the corresponding intention purpose is "how to do", and at this time, answer information corresponding to the intention object and the intention purpose is acquired from the knowledge base, and the answer information is fed back to the user.
The knowledge base is a knowledge source and basis for acquiring answers corresponding to the questions in the man-machine conversation process.
It should be noted that, during the process of the man-machine conversation, there are various ways in which the terminal device feeds back the answer information to the user, for example, the terminal device may play the answer information by voice, or the answer information fed back to the user is displayed on the man-machine interaction interface of the terminal device.
For another example, the information currently input by the user is "which submarine scoop nearest to home", the "home" in the information can be determined as an intention object whose information is unclear by analysis, and the purpose of analyzing the "home" input by the user is to expect that the terminal device obtains the submarine scoop nearest to the home address of the user according to the home address of the user in combination with the intention purpose of the information.
In the third example, target information is analyzed to obtain the hidden intention of a user; acquiring related auxiliary information from an exclusive corpus of a user according to the hidden intention; calling a related application program to perform information processing according to the intention object, the intention purpose and the auxiliary information in the target information; and obtaining a response result corresponding to the hiding intention and feeding back the response result to the user.
The auxiliary information includes personalized information of the user, wherein the personalized information may include, but is not limited to, preference information, habit information, and the like of the user, for example, the preference information may include which restaurant the user likes, taste preference, and the like, for example, the habit information may include getting-up time, travel habit information, and the like, and the implementation does not limit the auxiliary information.
For example, if the complete target information composed of the filling information and the currently input information is "the social security nearest to a certain cell" and the hidden intention of the user is to go to the social security by analyzing the target information, the trip mode preferred by the user is assumed to be the auxiliary information queried from the exclusive corpus of the user, at this time, according to the intention object and the intention purpose in the target information and the auxiliary information, the relevant map application program can be invoked to determine the social security nearest to the certain cell and determine the time required for the certain cell to trip to the nearest social security, and the response result provided for the user can be "the social security nearest to the user is at XXX, and likes to trip at ordinary times, and you only take five minutes in the past".
According to the natural language processing method, the missing intention object and/or the intention purpose information is completed through the historical corpus information in the exclusive corpus corresponding to the user, the intention object and intention purpose target information is generated, and corresponding processing is carried out according to the intention object and the intention purpose in the target information, so that integration of speech data with uncertain expression and intention is achieved based on the exclusive corpus of the user, the response efficiency of man-machine interaction can be improved, and the user experience degree of man-machine interaction is improved.
Based on the above embodiment, in order to facilitate the subsequent personalized filling of the missing intention objects and/or the intention destination information by combining with the exclusive corpus of the user, in an embodiment of the present invention, historical information of the user in the human-computer conversation may also be obtained, and the exclusive corpus of the user may be constructed based on the historical information.
As shown in fig. 2, the method may further include:
The preset corpus structure is used for screening the intention object and/or the intention purpose.
The preset corpus structure may include a corpus structure of user basic information, a corpus structure of user preferences, and a corpus structure corresponding to the transaction execution subject.
For the corpus structure of the basic user information, the corpus structure of the basic user information may include the corpus structures of the name, sex, home address, child condition, and the like of the user.
For the corpus structure preferred by the user, the corpus structure of the basic information of the user may include corpus structures such as user taste preference, travel preference, dressing preference, and the like.
For example, assuming that the history information input by the user is "my home is in a XXX cell", the historical information is extracted to determine that an intention object in the history information is the "XXX cell", and the intention object in the history information can be determined to conform to the corpus structure of the basic information of the user through analysis, and the XXX cell can be stored in the special corpus structure corresponding to the user according to the corpus structure of the basic information of the user.
For example, assuming that the history information input by the user is "how to do the social security card", it is possible to extract that the intention object in the history information is "social security card" and the corresponding intention purpose is "how to do", and in this case, it is possible to detect that the intention purpose of the intention object satisfies the corpus structure corresponding to the transaction executor, and store the intention object "social security card" and the intention purpose "how to do" in the corpus dedicated to the user in accordance with the corpus structure corresponding to the transaction executor.
And step 204, generating a recombination problem matched with the intention object and/or the intention purpose according to the history corpus information currently stored in the exclusive corpus corresponding to the user.
For example, assuming that the history information input by the user is "what to do", the intent purpose in the history information may be extracted as "what to do", and since the intent purpose "what to do" does not conform to the preset corpus structure, at this time, the reformulation question corresponding to the intent purpose may be generated according to the history corpus information currently stored in the exclusive corpus corresponding to the user, and the generated reformulation question may be "what to do".
And step 205, outputting the reorganization question to a user, and acquiring filling response information which is input by the user and corresponds to the reorganization question.
In different application scenarios, the output manner of the recombination problem to the user is different, for example, the recombination problem may be output to the user through a human-computer interaction interface, or the recombination problem may be output to the user in a voice broadcast manner, which is not limited in this embodiment.
And step 206, extracting the intention object and/or intention purpose meeting the preset corpus structure according to the recombination problem and the filling response information, and storing the intention object and/or intention purpose in the exclusive corpus corresponding to the user.
For example, assuming that the history information input by the user is "my home is east", the historical information is extracted to determine that the intention object in the history information is "my home is east", and the intention object in the history information is determined to be not in accordance with the corpus structure of the basic information of the user by analysis, and a reorganization problem of the intention object may be generated according to the history corpus information currently stored in the exclusive corpus corresponding to the user, the reorganization problem may be "where your family is specific", and it is assumed that the filling response information input by the user for the reorganization problem is "a certain cell". At this time, the intention object satisfying the preset corpus structure can be extracted according to the recombination problem and the filling response information, the intention object satisfying the preset corpus structure is a certain cell, and the certain cell is stored in the exclusive corpus corresponding to the user according to the corpus structure of the user basic information.
For example, assume that the historical information input by the user is "how long to drive a vehicle from home to Shenzhen university", wherein a schematic diagram of a human-computer interaction interface containing the information input by the user is shown in FIG. 3 a. After the information input by the user is obtained, by analyzing the information, it is determined that an intention object "home" with ambiguous information exists in the information, and it is determined whether the home address information of the user exists in an exclusive corpus corresponding to the user, if not, a reorganization problem matching the intention object is generated according to the history corpus information currently stored in the exclusive corpus, for example, the reorganization problem may be "ask where you are home", where a schematic diagram of a human-computer interaction interface including the reorganization problem is shown in fig. 3 b. After the user views the recombination problem in the human-computer interaction interface, the user inputs filling response information corresponding to the recombination problem, where a schematic diagram of the human-computer interaction interface including the filling response information is shown in fig. 3 c. And then, extracting an intention object 'science and science park' meeting the preset corpus structure according to the reorganization problem and the filling response information, and storing the home address information of the user in the exclusive corpus of the user, so that when the intention object 'home' with unclear information is missing in the information input again by the user, the home address information corresponding to the 'home' can be obtained from the exclusive corpus, and the home address information is filling information corresponding to the intention object 'home' with unclear information. Meanwhile, in order to respond to the user's question, a final question "how long to drive a vehicle from the science park to shenzhen university" can be obtained according to the home address information and the question input by the user, time information corresponding to the question is determined through a map application program, and a response result is provided to the user according to a certain structure, wherein a schematic diagram of a human-computer interaction interface including the response result is shown in fig. 3 d. Therefore, the user can conveniently obtain answers of the corresponding questions through human-computer interaction, and the efficiency of obtaining information by the user is improved.
In this example, the history dialogue information corresponding to the user and the mobile phone each time is analyzed, and the exclusive corpus corresponding to the user is continuously updated, so that a humanized exclusive corpus of the user can be formed, and further, the target information subsequently recombined according to the history corpus information in the exclusive corpus corresponding to the user has user pertinence, and meets the personalized requirements of the user.
Besides the above-mentioned method for constructing the exclusive corpus of the user, as a possible implementation method, the historical dialogue information of each man-machine dialogue of the user can be obtained, the intention object and the intention purpose in the historical dialogue information are extracted, the intention object and the intention purpose are clustered, and the clustering result is stored in the exclusive corpus of the user according to a certain structure, so as to form a structured exclusive corpus structure.
In order to more intuitively explain the implementation process of the natural language processing method according to the embodiment of the present invention, the following example is taken in conjunction with the application thereof in a specific scenario:
in this example, the terminal device is a smart phone, and the user performs a man-machine interaction in a voice manner.
The specific process is as follows: in the process of man-machine interaction between a user and a smart phone, the user can input voice information through a voice acquisition device of the smart phone, wherein a schematic diagram of a man-machine interaction interface corresponding to a voice input function of the smart phone is shown in fig. 4a, and a schematic diagram of a man-machine interaction interface containing text information corresponding to the voice information is assumed to be "which submarine scoop closest to home" in the voice information currently input by the user, as shown in fig. 4b, the user can be detected to have an ambiguous intention object "home" in the voice information currently input by the user, the intention in the voice information is to "obtain the submarine scoop closest to home", at this time, the user is determined to speak "home" according to the intention purpose in the voice information, and the terminal device is actually intended to obtain the submarine scoop closest to the home address of the user according to the home address of the user, so as to answer the problem of the user, at this time, the dedicated corpus of the user can be called, and the home address of the user is obtained from the dedicated corpus. Then, obtaining information of the seafloor fishing nearest to the home address of the user, and providing corresponding answer information to the user, wherein a schematic diagram of a human-computer interaction interface including answer information of machine response is shown in fig. 4c, where it is to be noted that in fig. 4c, the answer information only shows a name of the seafloor fishing nearest to the home address of the user, and in practical applications, other information of the seafloor fishing shop of the user, such as address information, introduction information, activity information, and the like, may also be provided to the user.
It should be noted that, the above description is only an example of displaying the dialog information between the user and the machine on the human-computer interaction interface of the smart phone, and in practical application, the smart device may also directly feed back the response result to the user in a voice playing manner, which is not limited in this embodiment.
In accordance with the natural language processing methods provided in the foregoing embodiments, an embodiment of the present invention further provides a natural language processing apparatus, and since the natural language processing apparatus provided in the embodiment of the present invention corresponds to the natural language processing methods provided in the foregoing embodiments, the embodiments of the foregoing natural language processing methods are also applicable to the natural language processing apparatus provided in the embodiment, and are not described in detail in the embodiment.
Fig. 5 is a schematic structural diagram of a natural language processing apparatus according to an embodiment of the present invention, as shown in fig. 5, the natural language processing apparatus includes a detection module 110, an acquisition module 120, an extraction module 130, and a processing module 140, wherein:
the detecting module 110 is configured to obtain information currently input by a user, and detect whether there is a missing intention object and/or intention destination in the information.
The intention object is a body to which the user inputs information to express the intention.
Wherein the intended purpose is the purpose that the user wants to achieve to input information.
It should be noted that the user may input information in various ways, for example, the user may input information in a text or voice manner.
It should be understood that, when the user inputs information in a text manner, the obtained information currently input by the user is text information, and when the user inputs information in a voice manner, the obtained information currently input by the user is voice information.
It should be noted that, in practical application, a user may select to input information in a text or voice manner according to a requirement, which is not limited in this embodiment.
The obtaining module 120 is configured to, if it is detected that there is a missing intention object and/or intention purpose in the information, invoke an exclusive corpus corresponding to the user to obtain the historical corpus information of the user.
And the extracting module 130 is configured to perform natural language analysis on the historical corpus information of the user, and extract filling information matched with the missing intention object and/or the intention purpose.
And the processing module 140 is configured to combine the filling information and the currently input information into complete target information, and perform corresponding processing according to an intention object and an intention purpose in the target information.
Further, the detection module 110 is specifically configured to: detecting that there is no intention object in the information, or there is only an intention object whose information is ambiguous; and/or detecting that the information is not intended for the purpose, or that only the intended purpose for which the information is ambiguous is present.
In different application scenarios, the obtaining module 120 invokes the dedicated corpus corresponding to the user to obtain the historical corpus information of the user in different manners, for example, as follows:
in an embodiment of the present invention, the obtaining module 120 is specifically configured to: acquiring a login account of a user; and calling the exclusive corpus corresponding to the login account to acquire the historical corpus information of the user.
In another embodiment of the present invention, the obtaining module 120 is specifically configured to: carrying out voiceprint processing on the currently input information to extract user voiceprint characteristics of the user; and calling a dedicated corpus corresponding to the voiceprint features of the user to acquire historical corpus information of the user.
In one embodiment of the present invention, in order to establish the dedicated corpus of the user, it is convenient to perform personalized filling on the missing intention objects and/or the intention destination information in subsequent combination with the dedicated corpus of the user. On the basis of the embodiment of the apparatus shown in fig. 5, as shown in fig. 6, the apparatus may further include:
a preprocessing module 150, configured to perform natural language processing on the history information input by the user each time, and extract an intention object and/or an intention purpose; detecting whether the intention object and/or the intention purpose meet a preset corpus structure or not; and if the fact that the preset corpus structure is met is detected, storing the intention object and/or the intention purpose in the exclusive corpus corresponding to the user according to the preset corpus structure.
In an embodiment of the present invention, the preprocessing module 150 is further configured to: if the situation that the preset corpus structure is not met is detected, generating a recombination problem matched with the intention object and/or the intention purpose according to the history corpus information currently stored in the exclusive corpus corresponding to the user; outputting the recombination problem to a user, and acquiring filling response information which is input by the user and corresponds to the recombination problem; and extracting an intention object meeting the preset corpus structure and/or an intention purpose according to the recombination problem and the filling response information, and storing the intention object and/or the intention purpose in an exclusive corpus corresponding to the user.
In different application scenarios, the processing module 140 performs corresponding processing according to the intention object and the intention purpose in the target information in different manners, which are illustrated as follows:
as an example, the processing module 140 is specifically configured to: analyzing the target information to obtain a processing instruction of a user; calling a related application program to perform information processing according to the intention object and the intention purpose in the target information; and obtaining a response result corresponding to the processing instruction and feeding back the response result to the user.
As another example, the processing module 140 is specifically configured to: analyzing the target information to obtain the hidden intention of the user; acquiring related query auxiliary information from an exclusive corpus of a user according to the hidden intention; calling a related application program to perform information processing according to the intention object, the intention purpose and the auxiliary information in the target information; and obtaining a response result corresponding to the hiding intention and feeding back the response result to the user.
According to the natural language processing device, the missing intention object and/or the intention purpose information is supplemented through the history corpus information in the exclusive corpus corresponding to the user, the intention object and intention purpose target information is generated, and corresponding processing is carried out according to the intention object and the intention purpose in the target information, so that integration of speech data with uncertain expression and intention is achieved based on the exclusive corpus of the user, the response efficiency of man-machine interaction can be improved, and the user experience degree of man-machine interaction is improved.
In order to implement the above embodiments, the present invention further provides a terminal device. Fig. 7 is a schematic interaction flow diagram of a natural language processing method according to an embodiment of the present invention, which is described by taking an example of storing a dedicated corpus of a user in a terminal device-side memory. The specific process of the natural language processing method at the terminal equipment side is as follows: a user inputs information in a man-machine interaction interface, a processor in terminal equipment acquires the information currently input by the user and detects whether the information has a missing intention object and/or an intention purpose; if the information is detected to have a missing intention object and/or an intention purpose, calling an exclusive corpus corresponding to the user from a server to acquire historical corpus information of the user, performing natural language analysis on the historical corpus information of the user, extracting the missing intention object and/or filling information matched with the intention purpose, combining the filling information and the currently input information into complete target information, performing corresponding processing according to the intention object and the intention purpose in the target information, and feeding back a processing result of the corresponding processing to the user.
To sum up, the terminal device according to the embodiment of the present invention completes missing intention objects and/or information of intention purposes through the history corpus information in the exclusive corpus corresponding to the user, generates target information including the intention objects and the intention purposes, and performs corresponding processing according to the intention objects and the intention purposes in the target information, thereby implementing integration of speech data with ambiguous expression and intention based on the exclusive corpus of the user, and further improving response efficiency of human-computer interaction and user experience of human-computer interaction.
In order to implement the above embodiments, the present invention further provides a storage medium for storing an application program, where the application program is configured to execute the natural language processing method according to any one of the embodiments of the present invention.
In the description of the present invention, it is to be understood that the terms "first", "second" and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
In the description of the specification, reference to the description of the term "one embodiment", "some embodiments", "an example", "a specific example", or "some examples", etc., means that a particular feature or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (12)
1. A natural language processing method, comprising:
acquiring information currently input by a user, and detecting whether the information has a missing intention object and/or an intention purpose;
if the information is detected to have a missing intention object and/or intention purpose, calling an exclusive corpus corresponding to the user to acquire historical corpus information of the user;
performing natural language analysis on the historical corpus information of the user, and extracting filling information matched with the missing intention object and/or intention purpose;
combining the filling information and the currently input information into complete target information;
analyzing the target information to obtain the hidden intention of the user, wherein the hidden intention comprises the personalized information of the user;
acquiring related auxiliary information from the exclusive corpus corresponding to the user according to the hiding intention;
calling a related application program to perform corresponding information processing according to the intention object and the intention purpose in the target information and the auxiliary information;
and obtaining a response result corresponding to the hiding intention and feeding back the response result to the user.
2. The method of claim 1, wherein the detecting of the presence of the missing intended object, and/or the intended purpose, of the information comprises:
detecting that the information has no intention object or only an intention object whose information is ambiguous;
and/or the presence of a gas in the gas,
it is detected that the information is not intended for the purpose or that only an intended purpose for which the information is ambiguous is present.
3. The method according to claim 1, wherein the invoking of the proprietary corpus corresponding to the user to obtain the historical corpus information of the user comprises:
acquiring a login account of the user;
and calling an exclusive corpus corresponding to the login account to acquire the historical corpus information of the user.
4. The method according to claim 1, wherein the invoking of the proprietary corpus corresponding to the user to obtain the historical corpus information of the user comprises:
carrying out voiceprint processing on the currently input information to extract user voiceprint characteristics of the user;
and calling a dedicated corpus corresponding to the user voiceprint features to acquire historical corpus information of the user.
5. The method of claim 1, further comprising:
carrying out natural language processing on the historical information input by the user each time, and extracting an intention object and/or an intention purpose;
detecting whether the intention object and/or the intention purpose meet a preset corpus structure or not;
and if the fact that the preset corpus structure is met is detected, storing the intention object and/or the intention purpose in an exclusive corpus corresponding to the user according to the preset corpus structure.
6. The method according to claim 5, wherein after said detecting whether said intended object, and/or said intended purpose, satisfies a preset corpus structure, further comprising:
if the situation that the preset corpus structure is not met is detected, generating a recombination problem matched with the intention object and/or the intention purpose according to the history corpus information currently stored in the exclusive corpus corresponding to the user;
outputting the recombination problem to the user, and acquiring filling response information which is input by the user and corresponds to the recombination problem;
and extracting an intention object meeting a preset corpus structure and/or an intention purpose according to the recombination problem and the filling response information, and storing the intention object and/or the intention purpose in an exclusive corpus corresponding to the user.
7. The method according to any one of claims 1-6, wherein the performing corresponding processing according to the intention object and the intention purpose in the target information comprises:
analyzing the target information to obtain a processing instruction of the user;
calling a related application program to perform information processing according to the intention object and the intention purpose in the target information;
and obtaining a response result corresponding to the processing instruction and feeding back the response result to the user.
8. A natural language processing apparatus, comprising:
the detection module is used for acquiring the information currently input by a user and detecting whether the information has a missing intention object and/or an intention purpose;
the acquisition module is used for calling an exclusive corpus corresponding to the user to acquire historical corpus information of the user if the information is detected to have a missing intention object and/or intention purpose;
the extraction module is used for carrying out natural language analysis on the historical corpus information of the user and extracting filling information matched with the missing intention object and/or intention purpose;
the processing module is used for combining complete target information according to the filling information and the currently input information; analyzing the target information to obtain the hidden intention of the user, wherein the hidden intention comprises the personalized information of the user; acquiring related auxiliary information from the exclusive corpus corresponding to the user according to the hiding intention; and calling a related application program to perform corresponding information processing according to the intention object and the intention purpose in the target information and the auxiliary information.
9. The apparatus of claim 8, wherein the processing module is further configured to:
if the situation that the preset corpus structure is not met is detected, generating a recombination problem matched with the intention object and/or the intention purpose according to the history corpus information currently stored in the exclusive corpus corresponding to the user;
outputting the recombination problem to the user, and acquiring filling response information which is input by the user and corresponds to the recombination problem;
and extracting an intention object meeting a preset corpus structure and/or an intention purpose according to the recombination problem and the filling response information, and storing the intention object and/or the intention purpose in an exclusive corpus corresponding to the user.
10. The apparatus according to any one of claims 8 to 9, wherein the processing module is specifically configured to:
analyzing the target information to obtain a processing instruction of the user;
calling a related application program to perform information processing according to the intention object and the intention purpose in the target information;
and obtaining a response result corresponding to the processing instruction and feeding back the response result to the user.
11. A terminal device, comprising:
memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the natural language processing method according to any one of claims 1 to 7.
12. A storage medium for storing an application program for executing the natural language processing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810434229.4A CN108664472B (en) | 2018-05-08 | 2018-05-08 | Natural language processing method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810434229.4A CN108664472B (en) | 2018-05-08 | 2018-05-08 | Natural language processing method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108664472A CN108664472A (en) | 2018-10-16 |
| CN108664472B true CN108664472B (en) | 2021-08-06 |
Family
ID=63778818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810434229.4A Active CN108664472B (en) | 2018-05-08 | 2018-05-08 | Natural language processing method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108664472B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210824B (en) * | 2018-11-21 | 2023-04-07 | 深圳绿米联创科技有限公司 | Voice information processing method and device, electronic equipment and storage medium |
| CN109346083A (en) * | 2018-11-28 | 2019-02-15 | 北京猎户星空科技有限公司 | A kind of intelligent sound exchange method and device, relevant device and storage medium |
| CN109857843B (en) * | 2018-12-25 | 2023-01-17 | 科大讯飞股份有限公司 | Interaction method and system based on document |
| CN110008413B (en) * | 2019-03-14 | 2023-11-10 | 海信集团有限公司 | Traffic travel problem query method and device |
| CN111858829A (en) * | 2019-04-30 | 2020-10-30 | 广东小天才科技有限公司 | A method and device for multi-intent sorting, and storage medium |
| CN110413250B (en) | 2019-06-14 | 2021-06-01 | 华为技术有限公司 | Voice interaction method, device and system |
| CN110798506B (en) * | 2019-09-27 | 2023-03-10 | 华为技术有限公司 | Method, device and equipment for executing command |
| CN111737972A (en) * | 2020-05-20 | 2020-10-02 | 华为技术有限公司 | Method and device for realizing natural language understanding in human-computer interaction system |
| CN111611369B (en) * | 2020-05-22 | 2023-06-20 | 腾讯科技（深圳）有限公司 | Interaction method and related device based on artificial intelligence |
| CN111738018A (en) * | 2020-06-24 | 2020-10-02 | 深圳前海微众银行股份有限公司 | An intent understanding method, apparatus, device and storage medium |
| CN111833872B (en) * | 2020-07-08 | 2021-04-30 | 北京声智科技有限公司 | Voice control method, device, equipment, system and medium for elevator |
| CN116308676A (en) * | 2023-03-30 | 2023-06-23 | 中国工商银行股份有限公司 | Quotation method, quotation device, quotation apparatus, quotation storage medium, and quotation program product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106357942A (en) * | 2016-10-26 | 2017-01-25 | 广州佰聆数据股份有限公司 | Intelligent response method and system based on context dialogue semantic recognition |
| CN106503046A (en) * | 2016-09-21 | 2017-03-15 | 北京光年无限科技有限公司 | Exchange method and system based on intelligent robot |
| CN106777018A (en) * | 2016-12-08 | 2017-05-31 | 竹间智能科技（上海）有限公司 | To the optimization method and device of read statement in a kind of intelligent chat robots |
| CN107533542A (en) * | 2015-01-23 | 2018-01-02 | 微软技术许可有限责任公司 | Method for understanding incomplete natural language queries |
-
2018
- 2018-05-08 CN CN201810434229.4A patent/CN108664472B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107533542A (en) * | 2015-01-23 | 2018-01-02 | 微软技术许可有限责任公司 | Method for understanding incomplete natural language queries |
| CN106503046A (en) * | 2016-09-21 | 2017-03-15 | 北京光年无限科技有限公司 | Exchange method and system based on intelligent robot |
| CN106357942A (en) * | 2016-10-26 | 2017-01-25 | 广州佰聆数据股份有限公司 | Intelligent response method and system based on context dialogue semantic recognition |
| CN106777018A (en) * | 2016-12-08 | 2017-05-31 | 竹间智能科技（上海）有限公司 | To the optimization method and device of read statement in a kind of intelligent chat robots |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108664472A (en) | 2018-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108664472B (en) | Natural language processing method, device and equipment | |
| CN112365892B (en) | Man-machine conversation method, device, electronic device and storage medium | |
| CN108446322B (en) | Method and device for realizing intelligent question-answering system | |
| US10438586B2 (en) | Voice dialog device and voice dialog method | |
| CN110335595A (en) | Speech recognition-based interrogation dialogue method, device and storage medium | |
| US9271111B2 (en) | Response endpoint selection | |
| CN111106995B (en) | Message display method, device, terminal and computer readable storage medium | |
| US11062708B2 (en) | Method and apparatus for dialoguing based on a mood of a user | |
| CN113132214B (en) | Dialogue method, dialogue device, dialogue server and dialogue storage medium | |
| WO2020253064A1 (en) | Speech recognition method and apparatus, and computer device and storage medium | |
| CN110610698B (en) | Voice labeling method and device | |
| CN112364622B (en) | Dialogue text analysis method, device, electronic device and storage medium | |
| CN110019687A (en) | A kind of more intention assessment systems, method, equipment and the medium of knowledge based map | |
| CN109271503A (en) | Intelligent answer method, apparatus, equipment and storage medium | |
| CN107704612A (en) | Dialogue exchange method and system for intelligent robot | |
| CN108519998B (en) | Problem guiding method and device based on knowledge graph | |
| CN110580516A (en) | interaction method and device based on intelligent robot | |
| CN113590776A (en) | Text processing method and device based on knowledge graph, electronic equipment and medium | |
| CN108877794A (en) | For the method, apparatus of human-computer interaction, electronic equipment and computer readable storage medium | |
| CN110442867A (en) | Image processing method, device, terminal and computer storage medium | |
| CN115098655A (en) | Common question answering method, system, equipment and medium | |
| CN108897517B (en) | Information processing method and electronic equipment | |
| CN112579031A (en) | Voice interaction method and system and electronic equipment | |
| CN107958059A (en) | Intelligent answer method, apparatus, terminal and computer-readable recording medium | |
| CN111353290B (en) | Method and system for automatically responding to user inquiry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |